Advertisement

Continuous-flow ventricular assist device exchange is safe and effective in prolonging support time in patients with end-stage heart failure

Open ArchivePublished:September 10, 2014DOI:https://doi.org/10.1016/j.jtcvs.2014.08.054

      Objective

      Although the development of continuous-flow ventricular assist devices (CF-VAD) has improved the reliability of these devices, VAD exchange is still occasionally necessary. The focus of this study was to analyze our institution's entire experience with primary CF-VAD implants, evaluate the baseline variables, determine which factors predict the need for exchange, and evaluate the impact of exchange on survival and event-free survival.

      Methods

      We retrospectively reviewed the data of all patients in a single center who received a primary CF-VAD implant between December 1999 and December 2013. All CF-VAD exchanges were reviewed; demographics, indications, preoperative and operative data, and clinical outcomes were summarized. Univariate and multivariable regression analyses were performed to ascertain predictors for exchange. Time-to-event and survival analyses were also performed.

      Results

      We identified 469 patients who underwent 546 CF-VAD implantations. Of these patients, 66 (14%) underwent 77 exchanges from one CF-VAD to another. The primary indications included hemolysis or thrombosis (n = 49; 63.6%), infection (n = 9; 11.7%), or other causes (n = 19; 24.7%). Survival was not significantly different between the exchange and nonexchange groups. Multivariable regression analysis identified a history of cerebrovascular events as a significant predictor for exchange. Among exchange patients, 11 underwent heart transplantation, 3 had their CF-VADs explanted, 26 had ongoing support, and 26 died during device support.

      Conclusions

      In our series of contemporary CF-VAD exchanges, a history of previous cerebrovascular events was a significant predictor for exchange. Exchange did not affect early or late survival. Our data suggest that aggressive surgical treatment of pump-related complications with exchange is safe and justified.

      CTSNet classification

      Abbreviations and Acronyms:

      CI (confidence interval), CF-VAD (continuous-flow ventricular assist device), CPB (cardiopulmonary bypass), INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support), MCS (nechanical circulatory support), OR (odds ratio), SD (standard deviation), VAD (ventricular assist device)
      See related commentary on pages 279-80.
      The use of continuous-flow ventricular assist devices (CF-VADs) has had a major impact on our ability to successfully treat end-stage cardiac disease. CF-VADs can be used as a bridge to transplantation, a bridge to recovery, or as destination therapy with meaningful clinical results.
      • Farrar D.J.
      • Holman W.R.
      • McBride L.R.
      • Kormos R.L.
      • Icenogle T.B.
      • Hendry P.J.
      • et al.
      Long-term follow-up of Thoratec ventricular assist device bridge-to-recovery patients successfully removed from support after recovery of ventricular function.
      • Frazier O.H.
      • Myers T.J.
      Left ventricular assist system as a bridge to myocardial recovery.
      • Aaronson K.D.
      • Slaughter M.S.
      • Miller L.W.
      • McGee E.C.
      • Cotts W.G.
      • Acker M.A.
      • et al.
      Use of an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation.
      • Birks E.J.
      • George R.S.
      • Hedger M.
      • Bahrami T.
      • Wilton P.
      • Bowles C.T.
      • et al.
      Reversal of severe heart failure with a continuous-flow left ventricular assist device and pharmacological therapy: a prospective study.
      • Kirklin J.K.
      • Naftel D.C.
      • Kormos R.L.
      • Stevenson L.W.
      • Pagani F.D.
      • Miller M.A.
      • et al.
      Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients.
      • Pagani F.D.
      • Miller L.W.
      • Russell S.D.
      • Aaronson K.D.
      • John R.
      • Boyle A.J.
      • et al.
      Extended mechanical circulatory support with a continuous-flow rotary left ventricular assist device.
      • Miller L.W.
      • Pagani F.D.
      • Russell S.D.
      • John R.
      • Boyle A.J.
      • Aaronson K.D.
      • et al.
      Use of a continuous-flow device in patients awaiting heart transplantation.
      • Park S.J.
      • Milano C.A.
      • Tatooles A.J.
      • Rogers J.G.
      • Adamson R.M.
      • Steidley D.E.
      • et al.
      Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy.
      Over the past decade, mechanical circulatory support (MCS) devices have supported thousands of patients with excellent overall outcomes,
      • Kirklin J.K.
      • Naftel D.C.
      • Kormos R.L.
      • Stevenson L.W.
      • Pagani F.D.
      • Miller M.A.
      • et al.
      Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients.
      particularly compared with medical therapy alone.
      • Rose E.A.
      • Gelijns A.C.
      • Moskowitz A.J.
      • Heitjan D.F.
      • Stevenson L.W.
      • Dembitsky W.
      • et al.
      Long-term use of a left ventricular assist device for end-stage heart failure.
      The experience with this technology continues to rapidly expand; more than 12,000 patients have been enrolled in the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) since 2006 in the United States alone.
      In comparison with first-generation devices, CF-VADs perform with remarkably improved durability and overall clinical outcomes.
      • Holman W.L.
      • Naftel D.C.
      • Eckert C.E.
      • Kormos R.L.
      • Goldstein D.J.
      • Kirklin J.K.
      Durability of left ventricular assist devices: Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) 2006 to 2011.
      • Slaughter M.S.
      • Rogers J.G.
      • Milano C.A.
      • Russell S.D.
      • Conte J.V.
      • Feldman D.
      • et al.
      Advanced heart failure treated with continuous-flow left ventricular assist device.
      However, adverse events, such as hemolysis, bleeding, thrombosis, infection, stroke, and mechanical failure, continue to be important and potentially devastating problems associated with this therapy.
      • Kirklin J.K.
      • Naftel D.C.
      • Kormos R.L.
      • Stevenson L.W.
      • Pagani F.D.
      • Miller M.A.
      • et al.
      Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients.
      • Cowger J.A.
      • Romano M.A.
      • Shah P.
      • Shah N.
      • Mehta V.
      • Haft J.W.
      • et al.
      Hemolysis: a harbinger of adverse outcome after left ventricular assist device implant.
      • Boyle A.J.
      • Jorde U.P.
      • Sun B.
      • Park S.J.
      • Milano C.A.
      • Frazier O.H.
      • et al.
      Pre-operative risk factors of bleeding and stroke during left ventricular assist device support: an analysis of more than 900 HeartMate II outpatients.
      • Zierer A.
      • Melby S.J.
      • Voeller R.K.
      • Guthrie T.J.
      • Ewald G.A.
      • Shelton K.
      • et al.
      Late-onset driveline infections: the Achilles’ heel of prolonged left ventricular assist device support.
      • Kalavrouziotis D.
      • Tong M.Z.
      • Starling R.C.
      • Massiello A.
      • Soltesz E.
      • Smedira N.G.
      • et al.
      Percutaneous lead dysfunction in the HeartMate II left ventricular assist device.
      When the adverse event is associated with a pump-related problem, it can often be managed by expeditious device exchange. Several groups, including our own, have demonstrated that judicious device exchange can potentially overcome some of the catastrophic consequences of such ventricular assist device (VAD)-related adverse events.
      • Abicht T.
      • Gordon R.
      • Meehan K.
      • Stosor V.
      • McCarthy P.
      • McGee Jr., E.
      Complex HeartMate II infection treated with pump exchange to HeartWare HVAD.
      • Moazami N.
      • Milano C.A.
      • John R.
      • Sun B.
      • Adamson R.M.
      • Pagani F.D.
      • et al.
      Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality.
      • Ota T.
      • Yerebakan H.
      • Akashi H.
      • Takayama H.
      • Uriel N.
      • Colombo P.C.
      • et al.
      Continuous-flow left ventricular assist device exchange: clinical outcomes.
      • Sajjad M.
      • Butt T.
      • Oezalp F.
      • Siddique A.
      • Wrightson N.
      • Crawford D.
      • et al.
      An alternative approach to explantation and exchange of the HeartWare left ventricular assist device.
      • Stulak J.M.
      • Cowger J.
      • Haft J.W.
      • Romano M.A.
      • Aaronson K.D.
      • Pagani F.D.
      Device exchange after primary left ventricular assist device implantation: indications and outcomes.
      • Adamson R.M.
      • Dembitsky W.P.
      • Baradarian S.
      • Chammas J.
      • Jaski B.
      • Hoagland P.
      • et al.
      HeartMate left ventricular assist system exchange: results and technical considerations.
      • Cohn W.E.
      • Mallidi H.R.
      • Frazier O.H.
      Safe, effective off-pump sternal sparing approach for HeartMate II exchange.
      • Jafar M.
      • Gregoric I.D.
      • Radovancevic R.
      • Cohn W.E.
      • McGuire N.
      • Frazier O.H.
      Urgent exchange of a HeartMate II left ventricular assist device after percutaneous lead fracture.
      • Gregoric I.D.
      Exchange techniques for implantable ventricular assist devices.
      The focus of this study was to analyze our institution's entire large experience with primary CF-VAD implants; to evaluate the baseline variables; and to determine which factors predict the need for exchange and the impact of exchange on survival and event-free survival.

      Methods

       Study Cohort

      Institutional Review Board approval with appropriate informed consent was obtained to perform a retrospective review of our center's patient database. All patients implanted with any CF-VAD between December 1999 and December 2013 were identified. Within this population, we identified those patients who underwent 1 or more VAD exchanges. Our aim was to compare the cohort of patients with VAD exchange with the cohort without exchange within the overall population of contemporary CF-VAD recipients. Any patient who had received a durable, non–CF-VAD at any point during their clinical course was excluded from this analysis.

       Primary End Points

      Demographics, indications, echocardiographic and hemodynamic parameters, operative, perioperative, and late clinical outcomes data were reviewed and summarized. Our primary aim was to determine significant independent predictors for VAD exchange. Secondary end points were overall survival and event-free survival while on VAD support and to the date of last follow-up. Event-free survival was defined as freedom from death, transplant, or VAD explant. The significance of VAD exchange as an independent predictor of survival was also evaluated.

       Statistical Analysis

      Demographics, indications for support, operative data, and clinical outcomes were compared between the CF-VAD exchange and nonexchange cohorts. Continuous variables were analyzed with the 2-sample t test, and categorical variables with the χ2 or Fisher exact test. Means are presented with standard deviations.
      Our primary aim of ascertaining predictors of VAD exchange was achieved using multivariable logistic regression modeling. We included all univariate variables that differed between the 2 cohorts (P < .05, or P < .1 trend), as well as all clinically relevant variables of interest. We did not include multiple correlated variables. Stepwise regression was performed eliminating all variables with P > .1. The result was a parsimonious clinically relevant model. This was performed for VAD exchange as the outcome variable, and repeated for overall survival and survival while on VAD support as the outcome, with VAD exchange remaining in the predictor model.
      Time-to-event survival analysis was performed creating Kaplan-Meier curves. Both the log-rank and the Wilcoxon tests were used to compare differences between the cohorts.
      Late follow-up was complete for 95% of patients for a mean time period of 2.1 ± 2.4 years (maximum, 13.3 years). There was minimal missing data (<5%) for all variables and/or data points. For all analyses, P values were 2-sided. Analyses were conducted with the R statistical software (Vienna, Austria).

      Results

       Patient Characteristics

      During our study period (December 1999 to December 2013), 469 patients underwent 546 CF-VAD implantations. Initially implanted pumps included the HeartMate II (Thoratec, Inc Pleasanton, Calif; n = 327), the Jarvik 2000 (Jarvik Heart Inc, New York, NY; n = 74), the HeartWare HVAD (HeartWare, Inc, Miami Lakes, Fla; n = 65), the DuraHeart (Terumo Heart, Inc, Ann Arbor, Mich; n = 2), and the MicroMed DeBakey (MicroMed Inc, Houston, Tex; n = 1). Of these patients, 66 underwent 77 VAD exchanges from their existing device to another CF-VAD; a 14% exchange incidence. The exchanged devices included the HeartMate II (n = 59), the Jarvik 2000 (n = 10), and the HeartWare HVAD (n = 8) devices. These devices were exchanged for the HeartMate II (n = 52), the Jarvik 2000 (n = 10), and the HeartWare HVAD (n = 15) devices. Total exchange rates for each device were 0.13 events per patient-year (15.5%) for the HeartMate II, 0.29 events per patient-year (11.9%) for the Jarvik 2000, and 0.16 events per patient-year (10%) for the HeartWare HVAD.
      Table 1 lists the baseline device, demographic, clinical, echocardiographic, and invasive hemodynamic data, and the operative and postoperative outcomes of the CF-VAD exchange cohort (n = 66 patients) and compares them with the corresponding data from the nonexchange cohort (n = 403 patients). There was a similar proportion of bridge to transplantation versus destination therapy indications in both groups. The group with 1 or more VAD exchanges had statistically longer duration of VAD support, more preoperative cerebrovascular events, higher platelet count and albumin levels, less INTERMACS class 1 status, and less frequently used preoperative temporary MCS (eg, Impella, Tandem Heart, intra-aortic balloon pump, and/or extracorporeal membrane oxygenation [ECMO]).
      Table 1Univariate analysis of baseline factors for nonexchange and exchange cohorts
      Total (n = 469)Nonexchange (n = 403)Exchange (n = 66)P value
      Device type, n (%).8
       Jarvik 200074 (15.8)64 (15.9)10 (15.1)
       HeartMate II327 (69.7)278 (69)49 (74.2)
       HeartWare HVAD65 (13.9)58 (14.4)7 (10.6)
       DuraHeart2 (0.4)2 (0.5)0
       MicroMed DeBakey1 (0.2)1 (0.2)0
      Age, y ± SD54.2 ± 13.954.4 ± 1451.3 ± 15.8.2
      Gender, n (%).07
       Male357 (76.1)313 (77.7)44 (66.7)
       Female112 (23.9)90 (22.3)22 (33.3)
      Race, n (%).8
       Native American3 (0.6)3 (0.7)0
       Asian10 (2.1)9 (2.2)1 (1.5)
       African American120 (25.6)99 (24.6)21 (31.8)
       White256 (54.6)222 (55.1)34 (51.5)
       Hispanic58 (12.4)50 (12.4)8 (12.1)
       Unknown/other22 (4.7)20 (5)2 (3)
      Device strategy, n (%).8
       Bridge to transplantation286 (61)247 (61.3)39 (59.1)
       Destination therapy183 (39)156 (38.7)27 (40.9)
      Cause of heart failure, n (%).1
       Nonischemic cardiomyopathy274 (58.4)229 (56.8)45 (68.2)
       Ischemic cardiomyopathy195 (41.6)174 (43.2)21 (31.8)
      Duration of support, d ± SD462.6 ± 502.4421.4 ± 481.2714.5 ± 556.7<.001
      Body mass index, kg/m2 ± SD27.7 ± 6.327.6 ± 6.328.4 ± 6.2.3
      Body surface area, m2 ± SD2 ± 0.32.0 ± 0.32.0 ± 0.3.3
      Comorbidities, n (%)
       Diabetes193 (41.2)169 (41.9)24 (36.4).5
       Hypertension223 (47.9)192 (48)31 (47)1.0
       Chronic obstructive pulmonary disease58 (12.4)48 (11.9)10 (15.4).6
       Stroke or transient ischemic attack49 (10.9)34 (8.8)15 (23).001
       Renal replacement therapy29 (6.2)29 (7.2)0.05
       Chronic kidney disease184 (39.3)159 (39.5)25 (37.9).9
       Respiratory failure68 (14.5)61 (15.2)7 (10.6).4
       Hypercoaguable state14 (3)10 (2.5)4 (6.1).2
       Deep venous thrombosis26 (5.5)23 (88.5)3 (11.5).9
       Pulmonary embolism22 (4.7)17 (4.2)5 (7.6).4
       Previous cardiac surgery170 (36.2)152 (37.7)18 (27.3).1
       Ventricular arrhythmia156 (33.3)131 (32.5)25 (37.9).5
       Atrial fibrillation130 (27.7)113 (28)17 (25.8).8
       Anticoagulation151 (34.5)134 (35.8)17 (26.6).2
       Antiplatelet therapy206 (47.1)181 (48.5)25 (39.1).2
       Intracardiac thrombus36 (7.8)31 (7.8)5 (7.6)1.0
       Infection52 (11.2)44 (11)8 (12.1)1.0
       Endocarditis3 (0.6)3 (0.7)01.0
       Active immunosuppression21 (4.5)15 (3.8)6 (9.1).1
       Tobacco use history195 (42.1)172 (43.3)23 (34.8).2
      Laboratory results
       Hemoglobin (g/dL)11.6 ± 2.111.6 ± 2.111.9 ± 2.2.1
       Leukocytes (K/μL)9.34 ± .39.4 ± 4.48.4 ± 3.2.1
       Platelets (K/μL)206 ± 87.4203.3 ± 88.2222 ± 81.1.04
       Sodium (mEq/L)135.1 ± 4.7135.1 ± 4.6135.1 ± 5.1.9
       Creatinine (mg/dL)1.4 ± 0.71.4 ± 0.71.4 ± 0.61.0
       Blood urea nitrogen (mg/dL)31.4 ± 18.931.8 ± 19.128.7 ± 17.7.2
       Aspartate aminotransferase (U/L)73.4 ± 173.977.3 ± 186.649.4 ± 44.3.4
       Alanine aminotransferase (U/L)79.8 ± 169.985.5 ± 181.745.5 ± 51.1
       Alkaline phosphatase (U/L)103.9 ± 56.9105.1 ± 56.697.5 ± 58.4.2
       Total bilirubin (mg/dL)2.0 ± 3.72.1 ± 3.91.4 ± 1.6.1
       Albumin (g/dL)3.7 ± 1.43.6 ± 1.43.9 ± 0.9.01
       Prealbumin (mg/dL)18.1 ± 8.217.9 ± 8.119 ± 8.5.7
       Prothrombin time (s)13.3 ± 5.513.4 ± 5.812.7 ± 3.8.2
       International normalized ratio1.2 ± 0.41.2 ± 0.41.2 ± 0.4.5
       Activated partial thromboplastin time (s)40.4 ± 22.340.5 ± 22.140.3 ± 22.9.7
       Lactate dehydrogenase (U/L)338.1 ± 205.1342 ± 210.5315.1 ± 169.1.1
       Brain natriuretic peptide (pg/mL)1155 ± 11271169.3 ± 1121.41083.2 ± 1165.7.5
      Echocardiogram, n (%)
       Left ventricular ejection fraction.7
      <20%281 (61.4)241 (61.3)40 (61.5)
      20%-29%148 (32.3)125 (31.8)23 (35.4)
      30%-39%27 (5.9)25 (6.4)2 (3)
      40%-49%2 (0.4)2 (0.5)0
       Left ventricular internal diastolic dimension6.7 ± 1.16.6 ± 1.16.7 ± 1.1.7
       Tricuspid regurgitation.5
      None/trace101 (23)84 (22.2)17 (27.4)
      Mild145 (33)127 (33.6)18 (29)
      Moderate156 (35.4)132 (34.9)24 (38.7)
      Severe38 (8.6)35 (9.3)3 (4.9)
       Mitral regurgitation.4
      None/trace67 (15)53 (13.9)14 (21.5)
      Mild125 (28)107 (28)18 (27.7)
      Moderate175 (39.1)151 (39.5)24 (36.9)
      Severe80 (17.9)71 (18.6)9 (13.9)
       Aortic regurgitation.4
      None/trace329 (79.7)280 (78.7)49 (86)
      Mild62 (15)58 (16.2)4 (7)
      Moderate15 (3.6)12 (3.4)3 (5.3)
      Severe2 (0.5)2 (0.6)0
      Not recorded5 (1.2)4 (1.1)1 (1.7)
       Right ventricular ejection fraction.2
      Normal111 (25.8)100 (27)11 (18.3)
      Depressed319 (74.2)270 (73)49 (81.7)
      Right heart catheterization
       Cardiac index1.9 ± 0.61.9 ± 0.61.9 ± 0.6.7
       Central venous pressure, mm Hg (mean)12.1 ± 7.612.3 ± 7.710.9 ± 6.6.3
       Right ventricular systolic pressure, mm Hg (mean)49.5 ± 15.949.3 ± 15.150.2 ± 19.91.0
       Pulmonary arterial pressure, mm Hg (mean)35.8 ± 11.335.7 ± 11.236.9 ± 11.5.4
       Pulmonary capillary wedge pressure, mm Hg25 ± 1024.8 ± 10.126.1 ± 10.1.3
       Pulmonary vascular resistance, Wood units ± SD3.6 ± 2.73.6 ± 2.83.5 ± 2.2.8
      Temporary mechanical circulatory support, n (%)
       Intra-aortic balloon pump208 (44.4)187 (46.4)21 (31.8).04
       Tandem heart59 (12.6)56 (13.9)3 (4.5).05
       Extracorporeal membrane oxygenation3 (0.6)3 (0.75)01.0
       Impella16 (3.4)15 (3.7)1 (1.5).6
      Intravenous drugs, n (%)
       Inotropes402 (85.7)344 (85.4)58 (87.9).7
       Vasopressors71 (15.1)65 (16.1)6 (9.1).2
      INTERMACS profile, n (%).005
       Profile 1 (crash and burn)83 (18.4)80 (20.7)3 (4.7)
       Profile 2 (sliding on inotropes)167 (37)144 (37.2)23 (35.9)
       Profile 3 (dependent stability)149 (33)118 (30.5)31 (48.4)
       Profile 4 (resting symptoms)52 (11.6)45 (11.6)7 (10.9)
      Intraoperative variables, n (%)
       Previous sternotomy165 (35.3)147 (36.7)18 (27.3).2
       Surgical approach1.0
      Median sternotomy401 (85.5)344 (85.4)57 (86.3)
      Nonsternotomy68 (14.5)59 (14.4)9 (13.7)
       Concomitant cardiovascular procedures181 (38.7)158 (39.3)23 (34.8).6
       Cardiopulmonary bypass use.8
      On-pump426 (91)365 (90.8)61 (92.4)
      Off-pump42 (9)37 (9.2)5 (7.6)
       Cardiopulmonary bypass time, minutes ± SD88.4 ± 51.388.7 ± 52.186.6 ± 46.5.8
       Delayed closure218 (47.1)189 (47.5)29 (44.6).7
       Right ventricular assist device18 (3.9)18 (4.5)0.2
      30-d mortality, n (%)55 (12)52 (13)3 (5).08
      Late outcomes, n (%)
       Death210168 (42)42 (64).001
       Transplant117106 (91)11 (9).1
       Explanted device2017 (85)3 (15)1.0
      Survival (VAD to last follow-up), y ± SD (range)2.1 ± 2.4 (0-13.3)2.0 ± 2.4 (0-13.3)2.4 ± 2.0 (0.02-9.2).2
      Continuous data are shown as means ± standard deviation (SD) and categorical data as numbers (%). INTERMACS, Interagency Registry for Mechanically Assisted Circulatory Support; VAD, ventricular assist device; SD, standard deviation.

       Exchange Cohort

      Table 2 summarizes the characteristics of the 66 patients who underwent 1 or more VAD exchanges and the rates of exchange by device. Hemolysis/suspected thrombosis was the primary indication (64%). Surgical exchange approaches were left subcostal (52%), resternotomy (40%), or left thoracotomy (8%).
      Table 2Descriptive characteristics of the exchange population
      First exchange (n = 66)Two or more exchanges (n = 11)Total (n = 77)
      Device type, n (%)
       Jarvik 20009 (13.7)0 (0)9 (11.7)
       HeartMate II49 (74.2)9 (82)58 (75.3)
       HeartWare HVAD8 (12.1)2 (18)10 (13)
      Indication for exchange, n (%)
       Thrombosis/hemolysis44 (66.7)5 (45.5)49 (63.6)
       Infection7 (10.6)2 (18.2)9 (11.7)
       Other15 (22.7)4 (36.3)19 (24.7)
      Mean time to exchange, d ± SD (range)339 ± 390 (1-1652)525 ± 394 (9-1259)375 ± 400 (1-1652)
      Surgical approach, n (%)
       Left subcostal36 (54.5)4 (36.4)40 (51.9)
       Resternotomy25 (37.9)6 (54.5)31 (40.3)
       Left thoracotomy5 (7.6)1 (9.1)6 (7.8)
      Continuous data are shown as means ± standard deviation and categorical data as numbers (%). SD, Standard deviation.

       Survival Analysis

      Figure 1 shows a Kaplan-Meier analysis of the overall outcomes between exchanged and nonexchanged patients. Figure 1, A and B, shows the overall and event-free survival, respectively, after primary VAD implant (to the date of last follow-up). Overall survival and event-free survival were no different up to 4 years.
      Figure thumbnail gr1
      Figure 1Kaplan-Meier analysis of outcomes between patients in whom a VAD was exchanged and those with no exchange. A, Survival after VAD implant (to most recent follow-up). B, Freedom from explant, transplant, or death after VAD implant (to most recent follow-up). C, Survival on VAD support. D, Freedom from explant, transplant, or death on VAD support. VAD, Ventricular assist device.
      Figure 1, C and D, shows the overall and event-free survival, respectively, during VAD support (time of primary VAD implant to date of death, explant, transplant, or last follow-up with ongoing support). Overall survival during VAD support was no different up to 4 years. Event-free survival during VAD support was greater in the exchange group by 4 years.
      Figure 2 illustrates overall survival by exchange indication. Although the results of the infection and other categories come from small samples, which preclude meaningful conclusions, the hemolysis and/or suspected thrombosis cohort showed a 2-year survival of 50%; this is comparable with survival in the general CF-VAD population.
      Figure thumbnail gr2
      Figure 2Survival (primary implant to most recent follow-up) by exchange indication.

       Multivariable Analysis

      Multivariable regression analysis with VAD exchange as the outcome showed previous cerebrovascular events (odds ratio [OR], 4.8; 95% confidence interval [CI], 2.2-10.2; P < .001) and temporary preoperative MCS (Impella, Tandem Heart, intra-aortic balloon pump, and/or ECMO) (OR, 0.2; 95% CI, 0.1-0.6; P = .001) to be the only significant, independent predictors of survival (Table 3).
      Table 3Multivariable logistic regression showing significant independent predictors of exchange and survival
      Odds ratio95% Confidence intervalP value
      Predictors of exchange
       Previous cerebrovascular event4.82.2-10.2<.001
       Preoperative mechanical circulatory support0.20.1-0.6.001
      Predictors of overall survival
       Exchange2.31.3-4.0.005
      Predictors of survival while on VAD support
       Exchange1.00.4-2.1.9
      VAD, Ventricular assist device.
      When overall survival was examined as the outcome, the impact of VAD exchange was significant (OR, 2.3; 95% CI, 1.3-4.0; P = .005). When freedom from an event (transplant, explant, or death) while on VAD support was the outcome, VAD exchange was not significant.

      Discussion

      The use of CF-VADs has dramatically increased in recent years.
      • Kirklin J.K.
      • Naftel D.C.
      • Kormos R.L.
      • Stevenson L.W.
      • Pagani F.D.
      • Miller M.A.
      • et al.
      Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients.
      As more patients are supported with these devices, a natural increase in the incidence of adverse events is expected and must be effectively managed. Although the durability of MCS devices has improved in the current era,
      • Holman W.L.
      • Naftel D.C.
      • Eckert C.E.
      • Kormos R.L.
      • Goldstein D.J.
      • Kirklin J.K.
      Durability of left ventricular assist devices: Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) 2006 to 2011.
      adverse events such as infection,
      • Zierer A.
      • Melby S.J.
      • Voeller R.K.
      • Guthrie T.J.
      • Ewald G.A.
      • Shelton K.
      • et al.
      Late-onset driveline infections: the Achilles’ heel of prolonged left ventricular assist device support.
      • Goldstein D.J.
      • Naftel D.
      • Holman W.
      • Bellumkonda L.
      • Pamboukian S.V.
      • Pagani F.D.
      • et al.
      Continuous-flow devices and percutaneous site infections: clinical outcomes.
      • Nienaber J.J.
      • Kusne S.
      • Riaz T.
      • Walker R.C.
      • Baddour L.M.
      • Wright A.J.
      • et al.
      Clinical manifestations and management of left ventricular assist device-associated infections.
      hemolysis,
      • Cowger J.A.
      • Romano M.A.
      • Shah P.
      • Shah N.
      • Mehta V.
      • Haft J.W.
      • et al.
      Hemolysis: a harbinger of adverse outcome after left ventricular assist device implant.
      • Ravichandran A.K.
      • Parker J.
      • Novak E.
      • Joseph S.M.
      • Schilling J.D.
      • Ewald G.A.
      • et al.
      Hemolysis in left ventricular assist device: a retrospective analysis of outcomes.
      thrombosis,
      • Kirklin J.K.
      • Naftel D.C.
      • Kormos R.L.
      • Pagani F.D.
      • Myers S.L.
      • Stevenson L.W.
      • et al.
      Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) analysis of pump thrombosis in the HeartMate II left ventricular assist device.
      • Starling R.C.
      • Moazami N.
      • Silvestry S.C.
      • Ewald G.
      • Rogers J.G.
      • Milano C.A.
      • et al.
      Unexpected abrupt increase in left ventricular assist device thrombosis.
      driveline injury,
      • Kalavrouziotis D.
      • Tong M.Z.
      • Starling R.C.
      • Massiello A.
      • Soltesz E.
      • Smedira N.G.
      • et al.
      Percutaneous lead dysfunction in the HeartMate II left ventricular assist device.
      • Jafar M.
      • Gregoric I.D.
      • Radovancevic R.
      • Cohn W.E.
      • McGuire N.
      • Frazier O.H.
      Urgent exchange of a HeartMate II left ventricular assist device after percutaneous lead fracture.
      and pump failure are still prevalent. In appropriately selected patients, these events can be effectively managed through primary VAD exchange.
      • Moazami N.
      • Milano C.A.
      • John R.
      • Sun B.
      • Adamson R.M.
      • Pagani F.D.
      • et al.
      Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality.
      • Ota T.
      • Yerebakan H.
      • Akashi H.
      • Takayama H.
      • Uriel N.
      • Colombo P.C.
      • et al.
      Continuous-flow left ventricular assist device exchange: clinical outcomes.
      • Stulak J.M.
      • Cowger J.
      • Haft J.W.
      • Romano M.A.
      • Aaronson K.D.
      • Pagani F.D.
      Device exchange after primary left ventricular assist device implantation: indications and outcomes.
      • Cohn W.E.
      • Mallidi H.R.
      • Frazier O.H.
      Safe, effective off-pump sternal sparing approach for HeartMate II exchange.
      • Potapov E.V.
      • Kaufmann F.
      • Stepanenko A.
      • Hening E.
      • Vierecke J.
      • Löw A.
      • et al.
      Pump exchange for cable damage in patients supported with HeartMate II left ventricular assist device.
      • Tang G.H.
      • Kim M.C.
      • Pinney S.P.
      • Anyanwu A.C.
      Failed repeated thrombolysis requiring left ventricular assist device pump exchange.
      In the present study, we report the largest single-center series of CF-VAD implants to date. Our experience with axial- and centrifugal-flow devices has included the Jarvik 2000, the HeartMate II, the HeartWare HVAD, the DuraHeart, and the MicroMed DeBakey assist devices. We conducted a retrospective analysis of the data of all patients receiving CF-VAD therapy, and, after excluding those who had received a pulsatile VAD at any time in their clinical course, we identified 469 patients who are the focus of this analysis.
      Univariate comparison of the data of the 66 patients who required an exchange with the data of the 403 patients who did not revealed few differences between the 2 groups. Preoperative baseline factors, including age, gender, race, indications for device implant, cause of heart failure, body mass index, and body surface area, were similar in both groups. The presence of comorbidities, including diabetes, hypertension, pulmonary disease, chronic kidney disease, hematologic disorders, arrhythmias, and tobacco use, was also similar. Analysis of preoperative laboratory data and echocardiographic and right heart catheterization measurements showed little to no statistically significant differences among these variables between groups. Other variables, such as the use of preoperative vasopressor and inotropic drugs and cardiopulmonary bypass (CPB), as well as CPB time, surgical approach, redo sternotomy, delayed sternal closure and emergency right VAD deployment, were all similar between groups.
      Univariate analysis showed that the differences in platelet count and albumin levels between the exchange and nonexchange groups were statistically significant. Increased platelet counts in the exchange group may indicate some degree of hypercoagulability, which may possibly lead to thrombosis and the need for device exchange. Rossi and colleagues
      • Rossi M.
      • Serraino G.F.
      • Jiritano F.
      • Renzulli A.
      What is the optimal anticoagulation in patients with a left ventricular assist device?.
      evaluated the ideal use of antiplatelet therapy in a best-practices literature review. Their report concluded that all patients with an axial-flow VAD should be on aspirin therapy alone or in combination with clopidogrel, in addition to warfarin. Our findings support that further preoperative hematologic evaluation in patients who are candidates for CF-VAD therapy may be warranted. In particular, evaluations assessing the patients' preoperative platelet functionality and responsiveness to antiplatelet therapy postoperatively may provide important insights into the risk of the patients experiencing hemolysis and/or suspected thrombosis during the period of device support.
      The exchange group experienced longer durations of durable VAD support. There was a trend toward increased 30-day mortality in the nonexchange group, which may have played a contributing factor. Although the value did not reach statistical significance, the proportion of patients on destination therapy was also greater in the exchange group. It can be assumed that fewer exchange patients were transplant candidates, thus lacking the option for urgent status 1A listing and heart transplantation.
      In a competing outcomes analysis (Figure E1), we demonstrate the overall outcomes for each patient. With the superimposed exchange curve, a nearly linear increase in the rate of exchange over time can be appreciated. The longer a patient has a pump, the more likely they are to require exchange. The patients on destination therapy are more likely to have a longer duration of support, as we have shown in our study. This puts them at a higher risk for exchange. Whereas bridge to transplant patients can be listed urgently as status 1A and are more likely to undergo transplantation, patients on destination therapy do not have this option. Therefore, in the setting of complications, exchange is often the best course of action. Our data show that exchange may be performed safely and is effective in extending the duration of support in patients who experience complications.
      Patients in the exchange group had higher rates of preoperative cerebrovascular events and lower rates of preoperative renal replacement therapy. Fewer exchange patients had a preoperative INTERMACS profile classification of 1 or required the use of temporary MCS devices, such as an intra-aortic balloon pump or percutaneous left VAD support. In our multivariable regression analysis for predictors of exchange, only 2 factors remained significant: the use of temporary MCS devices was negatively correlated and preoperative cerebrovascular event was strongly correlated with the need for CF-VAD exchange. Patients who required temporary MCS devices, including intra-aortic balloon pumps, ECMO, and the Impella and Tandem Heart percutaneous ventricular assist systems, were likely sicker and may not have survived long enough after implantation to have the need for CF-VAD exchange.
      The single positive predictor of CF-VAD exchange on multivariable regression was a preoperative cerebrovascular event, such as a transient ischemic attack or a cerebrovascular accident. This finding may be related to a potential hypercoaguable status in patients with a previous thromboembolic event. Alternatively, this finding may be related to a potentially poorer postoperative functional status, which has been shown to be associated with perioperative adverse outcomes in other forms of cardiac surgery.
      • Mayer C.
      • Ergina P.
      • Morin J.F.
      • Gold S.
      Self-reported functional status as a predictor of coronary artery bypass graft surgery outcome in elderly patients.
      Thrombosis, infection, driveline injury, and other potential factors for pump exchange could all be correlated with functional limitations after a cerebrovascular event. Our findings have resulted in the routine workup for all potential patients for VAD implantation with a history of a preoperative cerebrovascular accident to be modified to include a functionality assessment, as well as a preoperative hematology consultation investigating potential inherited or acquired hypercoaguable states.
      A closer analysis of the group of patients who required a pump exchange is summarized in Table 2. Our initial clinical experience with the Jarvik 2000 was described previously.
      • Frazier O.H.
      • Myers T.J.
      • Gregoric I.D.
      • Khan T.
      • Delgado R.
      • Croitoru M.
      • et al.
      Initial clinical experience with the Jarvik 2000 implantable axial-flow left ventricular assist system.
      The HeartMate II and the HVAD are the only approved CF-VADs in the United States at the time of this study. The indications for device exchange were hemolysis and/or thrombosis, infection, and other. In our study, we did not differentiate pump thrombosis from hemolysis because both are related events with multifactorial causes,
      • Uriel N.
      • Han J.
      • Morrison K.A.
      • Nahumi N.
      • Yuzefpolskaya M.
      • Garan A.R.
      • et al.
      Device thrombosis in HeartMate II continuous-flow left ventricular assist devices: a multifactorial phenomenon.
      which may be due to patient, pump, technical, or clinical management factors. The events grouped into the other category were driveline trauma, inlet malposition, pump failure of unknown cause, attempted suicide, and severe pain. Ota and colleagues
      • Ota T.
      • Yerebakan H.
      • Akashi H.
      • Takayama H.
      • Uriel N.
      • Colombo P.C.
      • et al.
      Continuous-flow left ventricular assist device exchange: clinical outcomes.
      recently reported similar results (19 of 30 devices [63%] were exchanged because of hemolysis or thrombosis), which are proportionally identical to our own results. HeartMate II registry data showed that 7 of 70 patients underwent pump exchange because of infection (8.9%), a result that is not greatly dissimilar from our own (11.7%).
      • Moazami N.
      • Milano C.A.
      • John R.
      • Sun B.
      • Adamson R.M.
      • Pagani F.D.
      • et al.
      Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality.
      Techniques described for exchanges have been reported previously by our own and other groups.
      • Sajjad M.
      • Butt T.
      • Oezalp F.
      • Siddique A.
      • Wrightson N.
      • Crawford D.
      • et al.
      An alternative approach to explantation and exchange of the HeartWare left ventricular assist device.
      • Cohn W.E.
      • Mallidi H.R.
      • Frazier O.H.
      Safe, effective off-pump sternal sparing approach for HeartMate II exchange.
      • Gregoric I.D.
      Exchange techniques for implantable ventricular assist devices.
      Nonsternotomy approaches included left subcostal incision and left anterior thoracotomy. The latter approach was used solely in patients implanted with a Jarvik 2000.
      Our analysis showed that the overall survival of the exchange group was not significantly different from that of the nonexchange group (Figure 1, A; P = .77). Similarly, Kaplan-Meier analysis of the overall freedom from event (transplant, explant, or death) showed no difference between groups (Figure 1, B; P = .11). When Kaplan-Meier analysis was repeated for VAD-supported survival, with follow-up censored by date of transplant, explant, or death, again there was no statistically significant difference between exchange and nonexchange groups (Figure 1, C; P = .21). On the other hand, the Kaplan-Meier analysis of freedom from event during VAD support showed a statistically significant increase in the event rate in the exchange group. There were more patients on destination therapy in the exchange cohort; likely more transplants and/or explants occurred in the nonexchange group.

       Limitations

      This study is limited by the retrospective nature of the data review. Furthermore, it presents the results obtained in a single center, which may not be representative of what is observed in other centers. Postoperative information including patient and device-related management factors and adverse event data were not available for all patients. However, this report is timely and relevant given the large sample size, multiple CF-VAD types with significant periods of follow-up, and the exchange incidence reported from a contemporary, high-volume center demonstrating good long-term outcomes in patients who require VAD exchange. Future multicenter prospective studies are warranted to better understand the predictors and natural history of CF-VAD exchanges.

      Conclusions

      In our large single-center experience with CF-VAD implants, preoperative variables including patient demographics, comorbidities, laboratory results, echocardiographic data, invasive catheterization, and operative data had little predictive power for CF-VAD exchange. The presence of cerebrovascular events before CF-VAD support did, however, have a strong correlation with the ultimate need for exchange. The decision for CF-VAD exchange should be considered for all patients who experience adverse events that can be managed appropriately by pump replacement. Exchange and nonexchange groups had similar overall survival and freedom from death while on support. Judicious and expeditious pump exchange is recommended in select patients.
      The authors would like to acknowledge Dr Ana Rodriguez for editorial services, as well as David Antoun, Rohan M. Shah, Ryan W. Fairchild, and Kelly Handy for their assistance with the data collection for this manuscript.

      Appendix

      Figure thumbnail fx1
      Figure E1Competing outcomes analysis of patients with a continuous-flow ventricular assist device (CF-VAD) with superimposed rate of exchange curve. This figure demonstrates the outcomes of all patients with CF-VAD in our study cohort. The survival curve inversely correlates with the death curve. The alive-on-support curve inversely correlates with the final event curves (transplant and explant). The exchange curve is superimposed to demonstrate the number of exchanges in this population over time.

      References

        • Farrar D.J.
        • Holman W.R.
        • McBride L.R.
        • Kormos R.L.
        • Icenogle T.B.
        • Hendry P.J.
        • et al.
        Long-term follow-up of Thoratec ventricular assist device bridge-to-recovery patients successfully removed from support after recovery of ventricular function.
        J Heart Lung Transplant. 2002; 21: 516-521
        • Frazier O.H.
        • Myers T.J.
        Left ventricular assist system as a bridge to myocardial recovery.
        Ann Thorac Surg. 1999; 68: 734-741
        • Aaronson K.D.
        • Slaughter M.S.
        • Miller L.W.
        • McGee E.C.
        • Cotts W.G.
        • Acker M.A.
        • et al.
        Use of an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation.
        Circulation. 2012; 125: 3191-3200
        • Birks E.J.
        • George R.S.
        • Hedger M.
        • Bahrami T.
        • Wilton P.
        • Bowles C.T.
        • et al.
        Reversal of severe heart failure with a continuous-flow left ventricular assist device and pharmacological therapy: a prospective study.
        Circulation. 2011; 123: 381-390
        • Kirklin J.K.
        • Naftel D.C.
        • Kormos R.L.
        • Stevenson L.W.
        • Pagani F.D.
        • Miller M.A.
        • et al.
        Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients.
        J Heart Lung Transplant. 2013; 32: 141-156
        • Pagani F.D.
        • Miller L.W.
        • Russell S.D.
        • Aaronson K.D.
        • John R.
        • Boyle A.J.
        • et al.
        Extended mechanical circulatory support with a continuous-flow rotary left ventricular assist device.
        J Am Coll Cardiol. 2009; 54: 312-321
        • Miller L.W.
        • Pagani F.D.
        • Russell S.D.
        • John R.
        • Boyle A.J.
        • Aaronson K.D.
        • et al.
        Use of a continuous-flow device in patients awaiting heart transplantation.
        N Engl J Med. 2007; 357: 885-896
        • Park S.J.
        • Milano C.A.
        • Tatooles A.J.
        • Rogers J.G.
        • Adamson R.M.
        • Steidley D.E.
        • et al.
        Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy.
        Circ Heart Fail. 2012; 5: 241-248
        • Rose E.A.
        • Gelijns A.C.
        • Moskowitz A.J.
        • Heitjan D.F.
        • Stevenson L.W.
        • Dembitsky W.
        • et al.
        Long-term use of a left ventricular assist device for end-stage heart failure.
        N Engl J Med. 2001; 345: 1435-1443
        • Holman W.L.
        • Naftel D.C.
        • Eckert C.E.
        • Kormos R.L.
        • Goldstein D.J.
        • Kirklin J.K.
        Durability of left ventricular assist devices: Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) 2006 to 2011.
        J Thorac Cardiovasc Surg. 2013; 146: 437-441.e1
        • Slaughter M.S.
        • Rogers J.G.
        • Milano C.A.
        • Russell S.D.
        • Conte J.V.
        • Feldman D.
        • et al.
        Advanced heart failure treated with continuous-flow left ventricular assist device.
        N Engl J Med. 2009; 361: 2241-2251
        • Cowger J.A.
        • Romano M.A.
        • Shah P.
        • Shah N.
        • Mehta V.
        • Haft J.W.
        • et al.
        Hemolysis: a harbinger of adverse outcome after left ventricular assist device implant.
        J Heart Lung Transplant. 2014; 33: 35-43
        • Boyle A.J.
        • Jorde U.P.
        • Sun B.
        • Park S.J.
        • Milano C.A.
        • Frazier O.H.
        • et al.
        Pre-operative risk factors of bleeding and stroke during left ventricular assist device support: an analysis of more than 900 HeartMate II outpatients.
        J Am Coll Cardiol. 2014; 63: 880-888
        • Zierer A.
        • Melby S.J.
        • Voeller R.K.
        • Guthrie T.J.
        • Ewald G.A.
        • Shelton K.
        • et al.
        Late-onset driveline infections: the Achilles’ heel of prolonged left ventricular assist device support.
        Ann Thorac Surg. 2007; 84: 515-520
        • Kalavrouziotis D.
        • Tong M.Z.
        • Starling R.C.
        • Massiello A.
        • Soltesz E.
        • Smedira N.G.
        • et al.
        Percutaneous lead dysfunction in the HeartMate II left ventricular assist device.
        Ann Thorac Surg. 2014; 97: 1373-1378
        • Abicht T.
        • Gordon R.
        • Meehan K.
        • Stosor V.
        • McCarthy P.
        • McGee Jr., E.
        Complex HeartMate II infection treated with pump exchange to HeartWare HVAD.
        ASAIO J. 2013; 59: 188-192
        • Moazami N.
        • Milano C.A.
        • John R.
        • Sun B.
        • Adamson R.M.
        • Pagani F.D.
        • et al.
        Pump replacement for left ventricular assist device failure can be done safely and is associated with low mortality.
        Ann Thorac Surg. 2013; 95: 500-505
        • Ota T.
        • Yerebakan H.
        • Akashi H.
        • Takayama H.
        • Uriel N.
        • Colombo P.C.
        • et al.
        Continuous-flow left ventricular assist device exchange: clinical outcomes.
        J Heart Lung Transplant. 2014; 33: 65-70
        • Sajjad M.
        • Butt T.
        • Oezalp F.
        • Siddique A.
        • Wrightson N.
        • Crawford D.
        • et al.
        An alternative approach to explantation and exchange of the HeartWare left ventricular assist device.
        Eur J Cardiothorac Surg. 2013; 43: 1247-1250
        • Stulak J.M.
        • Cowger J.
        • Haft J.W.
        • Romano M.A.
        • Aaronson K.D.
        • Pagani F.D.
        Device exchange after primary left ventricular assist device implantation: indications and outcomes.
        Ann Thorac Surg. 2013; 95 (discussion 1267-8): 1262-1267
        • Adamson R.M.
        • Dembitsky W.P.
        • Baradarian S.
        • Chammas J.
        • Jaski B.
        • Hoagland P.
        • et al.
        HeartMate left ventricular assist system exchange: results and technical considerations.
        ASAIO J. 2009; 55: 598-601
        • Cohn W.E.
        • Mallidi H.R.
        • Frazier O.H.
        Safe, effective off-pump sternal sparing approach for HeartMate II exchange.
        Ann Thorac Surg. 2013; 96: 2259-2261
        • Jafar M.
        • Gregoric I.D.
        • Radovancevic R.
        • Cohn W.E.
        • McGuire N.
        • Frazier O.H.
        Urgent exchange of a HeartMate II left ventricular assist device after percutaneous lead fracture.
        ASAIO J. 2009; 55: 523-524
        • Gregoric I.D.
        Exchange techniques for implantable ventricular assist devices.
        ASAIO J. 2008; 54: 14-19
        • Goldstein D.J.
        • Naftel D.
        • Holman W.
        • Bellumkonda L.
        • Pamboukian S.V.
        • Pagani F.D.
        • et al.
        Continuous-flow devices and percutaneous site infections: clinical outcomes.
        J Heart Lung Transplant. 2012; 31: 1151-1157
        • Nienaber J.J.
        • Kusne S.
        • Riaz T.
        • Walker R.C.
        • Baddour L.M.
        • Wright A.J.
        • et al.
        Clinical manifestations and management of left ventricular assist device-associated infections.
        Clin Infect Dis. 2013; 57: 1438-1448
        • Ravichandran A.K.
        • Parker J.
        • Novak E.
        • Joseph S.M.
        • Schilling J.D.
        • Ewald G.A.
        • et al.
        Hemolysis in left ventricular assist device: a retrospective analysis of outcomes.
        J Heart Lung Transplant. 2014; 33: 44-50
        • Kirklin J.K.
        • Naftel D.C.
        • Kormos R.L.
        • Pagani F.D.
        • Myers S.L.
        • Stevenson L.W.
        • et al.
        Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) analysis of pump thrombosis in the HeartMate II left ventricular assist device.
        J Heart Lung Transplant. 2014; 33: 12-22
        • Starling R.C.
        • Moazami N.
        • Silvestry S.C.
        • Ewald G.
        • Rogers J.G.
        • Milano C.A.
        • et al.
        Unexpected abrupt increase in left ventricular assist device thrombosis.
        N Engl J Med. 2014; 370: 33-40
        • Potapov E.V.
        • Kaufmann F.
        • Stepanenko A.
        • Hening E.
        • Vierecke J.
        • Löw A.
        • et al.
        Pump exchange for cable damage in patients supported with HeartMate II left ventricular assist device.
        ASAIO J. 2012; 58: 578-582
        • Tang G.H.
        • Kim M.C.
        • Pinney S.P.
        • Anyanwu A.C.
        Failed repeated thrombolysis requiring left ventricular assist device pump exchange.
        Catheter Cardiovasc Interv. 2013; 81: 1072-1074
        • Rossi M.
        • Serraino G.F.
        • Jiritano F.
        • Renzulli A.
        What is the optimal anticoagulation in patients with a left ventricular assist device?.
        Interact Cardiovasc Thorac Surg. 2012; 15: 733-740
        • Mayer C.
        • Ergina P.
        • Morin J.F.
        • Gold S.
        Self-reported functional status as a predictor of coronary artery bypass graft surgery outcome in elderly patients.
        Can J Cardiol. 2003; 19: 140-144
        • Frazier O.H.
        • Myers T.J.
        • Gregoric I.D.
        • Khan T.
        • Delgado R.
        • Croitoru M.
        • et al.
        Initial clinical experience with the Jarvik 2000 implantable axial-flow left ventricular assist system.
        Circulation. 2002; 105: 2855-2860
        • Uriel N.
        • Han J.
        • Morrison K.A.
        • Nahumi N.
        • Yuzefpolskaya M.
        • Garan A.R.
        • et al.
        Device thrombosis in HeartMate II continuous-flow left ventricular assist devices: a multifactorial phenomenon.
        J Heart Lung Transplant. 2014; 33: 51-59